The ongoing pattern of groundwater induced land subsidence in major valleys and agricultural regions of Iran has been recently documented by several studies (e.g. [1-4]) using C-band Interferometric Synthetic Aperture Radar (InSAR) observations. In this article we present the results of our research in which we evaluated the performance of C-band, L-band and Xband SAR data, using time-series method of small baseline subset (SBAS), to retrieve long time series of ground subsidence in agricultural regions in the country. Two major groundwater basins have been selected for this purpose: (1) Rafsanjan Valley in the Kerman province of central Iran and (2) Tehran Plain (capital of Iran). We also report on our experience using dualpolarimetry (HH/VV) X-band SAR data for Persistent Scatterer (PS) deformation analysis in natural terrains subject to high rate of deformation. 1. Land subsidence in Rafsanjan To investigate land subsidence in Rafsanjan we utilized 3 sets of images, including 12 Envisat ASAR data in a descending track covering March 2004–June 2006, 23 Envisat ASAR data in an ascending track covering June 2004–August 2007, and 10 ALOS PALSAR data in an ascending track covering December 2007–November 2010. The outline of the SAR frames has been illustrated in Fig. 1. We processed all SAR data with small baseline subset (SBAS) approach implemented in StaMPS software (http://homepages.see.leeds.ac.uk/~earahoo/stamps/inde x.html). Figure 1. Geographic location of the Rafsanjan valley in central Iran. Rectangles show the frames of radar sensors used in this study. Figure 2. Mean line-of-sight (LOS) velocity derived from Envisat ascending data. The black square is the reference area. COMPARISON OF X-BAND, L-BAND AND C-BAND RADAR IMAGES FOR MONITORING SUBSIDENCE IN AGRICULTURAL REGIONS Mahdi Motagh , Mahmud Haghshenas Haghighi , Roghaye Shamshiri , Mustapha Esmaeili (3) (1) GFZ German Research Center for Geosciences, Potsdam, Germany. Email: Motagh@gfz-potsdam.de (2) Norwegian University of Science and Technology, Norway. Email: roghayeh.shamshiri@ntnu.no (3) Department of surveying and Geomatics Engineering, University of Tehran, Iran. Email: mostafaesi@gmail.com _____________________________________ Proc. ‘Fringe 2015 Workshop’, Frascati, Italy 23–27 March 2015 (ESA SP-731, May 2015) Figure 3. Mean line-of-sight (LOS) velocity derived from Envisat descending data. Figure 4. Mean line-of-sight (LOS) velocity derived from ALOS ascending data Figs. 2-4 depict the mean line-of-sight (LOS deformation maps in the year 2004-2007 derived from Envisat ascending (Fig. 2) and Envisat descending data (Fig. 3) and in the year 2007-2010 derived from ALOS ascending data (Fig. 4). As seen in Figs. 2-4 the entire Rafsanjan valley has been affected by land subsidence with maximum rate of LOS motion reaching to 25 cm/year. The difference in the rate and pattern of LOS displacement maps between ascending and descending results can be attributed to the effect of horizontal motion. 2. Land subsidence in Tehran The analysis in this section has been divided into two parts. First we compare using SBAS approach timeseries results from different sensors to retrieve spatiotemporal pattern of land subsidence west of Tehran. The data include 39 C-band Envisat ASAR, 10 L-band ALOS PALSAR, and 15 X-band TerraSAR-X copolarized (VV) data. Secondly, we evaluate the use of polarimetric optimization method using dual-polarized data for Persistent Scatterer (PS) analysis in the same region. Our dual-polarimetry dataset includes 17 Xband SAR data (HH/VV) over southwest of Tehran acquired by the TerraSAR-X satellite between July 2013 and January 2014. 2.1 Multi-sensor InSAR survey Underground water exploitation in Tehran Plain has led to land subsidence in agricultural areas west of Tehran City [1,4]. Here we present the result of InSAR timeseries processing applied on a large dataset of X-band, C-band and L-band SAR imagery covering a long period of time between 2003 and 2013. Average velocity maps in the vertical direction, assuming no contribution from horizontal motion in the LOS direction, derived from different datasets are shown in Fig. 5. The spatial extent of the subsidence area is almost similar in the results obtained from different sensors (Fig. 5). However, densities of coherent pixels in TSX results are more than other datasets. Although short wavelength X-band SAR data are more sensitive to vegetation coverage and temporal decorrelation than C-band and L-band data, time interval between two consecutive TSX images are short enough (11 day) to derive coherent interferograms for deformation analysis in agricultural regions with large rate (> 20 cm/yr) of deformation. Density of coherent pixels in C-band Envisat results is less than other datasets. Figure 5. Average vertical velocity maps derived from different datasets for the subsidence bowl southwest of Tehran City. (a) Envisat 149, (b) Envisat 378, (c) Envisat 414, (d) ALOS, and (e) TSX. Black line shows the boundary of the urban and residential area in Tehran. 2.2 Dual-polarimetry analysis Using dual polarimetric SAR data we can produce for each pixel in a SAR image a scattering coefficient, μ , defined as

/pdf/comparison-of-x-band-l-band-and-c-band-radar-images-for-2l1pblh0gf.pdf

Mustapha Esmaeili

Papers

Comparison of X-band, L-band and C-band radar images for monitoring subsidence in agricultural regions